EXPANSION OF LIQUIDS. 



A singular correspondence has been observed, by M. Guy- 

 Lussac, between two particular liquids alcohol and sulphuret 

 of carbon, in the amount of their expansion by heat ; although 

 each of these liquids has a particular temperature at which it 

 boils 



Alcohol at 173 



Sulphuret of carbon at 1 1 6 



still the ratios of expansion from the addition, and of contrac- 

 tion from the loss of heat, are found to be uniformly the same 

 in the case of these two liquids, which are the only ones known 

 to possess such a relation. The number of liquids, however, 

 the expansions of which, under different degrees of heat, have 

 been examined, is exceedingly small ; although comparative ex- 

 periments may be made with much greater facility in regard to 

 li quids than solids. 



The second law is, that liquids are progressively more expan- 

 sible at higher than at lower temperatures. This is less the case 

 with mercury, perhaps, than with any other liquid. The expan- 

 sions of that liquid are, indeed, so uniform, as to render it ex- 

 tremely proper for the construction of the thermometer, as will 

 afterwards appear. The rate of expansion of mercury was 

 determined with extraordinary care by Dulong and Petit. 

 From lto 1 00 centigrade, mercury expands 1 measure on 

 1000,, 200 1 



200,, 300o n ?? n 1 53 



According to the same experimenters, the expansion of mer- 

 cury, confined in glass tubes, is only 1 on 64.8. The dilatation 

 of the glass causes the capacity of the instrument to be enlarged, 

 so that the whole expansion of the mercury is not indicated. 

 The only mode in which the error introduced by the expansion 

 of the enclosing 

 vessel can be 

 avoided, in ascer- 

 taining the ex- 

 pansions of li- 

 quids, is that 

 practised by Du- 

 long and Petit; 

 namely, heating 

 the liquid in one limb of a syphon, (see Figure) and observing 

 how high it rises above the level of the same liquid in the other 



